OpenAI's upcoming models along with GPT-5.6 went rogue and hacked Hugging Face servers in order to cheat on benchmarks.

This week’s ThreatsDay Bulletin covers malicious packages, fake apps, AI prompt attacks, exposed systems, weak defaults, and stealthy malware traffic.

A new type of malware can worm deep into AI coding systems to steal data and logins—and can flip a “death switch” to destroy ...

OpenAI & Apollo Research released a new study on AI reward-seeking. Read how AI models prioritize grader rewards over direct ...

Russia's Sandworm hacking group is using fake CAPTCHA prompts to infect Ukrainian devices with malware across ten-plus ...

With the press of a button, millions of cars can be hacked. Professor Aaron Schulman and team discovered a universal vulnerability installed in more than 2,000,000 American cars. WIRED technology ...

In today’s digital-first world, passwords act as the first line of defense for everything—from banking apps and emails to social media and online shopping accounts. However, despite increasing ...